CODEWITHSUNDEEP
pythonarraysnumpy
New2Coding
New2Coding

Reputation: 191

Numpy scale specific values of numpy array

I want to scale specific values (e.g. lager than 5) of an numpy array numbers by a multiplier (e.g. 2). I know I can achieve this by a loop, but I wanted to avoid loops. I think I can achieve this somehow with a numpy mask but I am not sure how to implement it. In order to demonstrate what I aim to achieve I used the imaginary function scale_array.

Here is a minimal working example

import numpy as np

numbers = np.array([-3, 5, 2, -1, -15, 10])
mask = np.abs(numbers) > 5

numbers_scaled = scale_array(array=numbers, mask=mask, scale_factor=2)
print(numbers_scaled)  # np.array([-3, 5, 2, -1, -30, 20])

Upvotes: 0

Views: 393

Answers (2)

PAb
PAb

Reputation: 561

As you write, mask = np.abs(numbers) > 5 gives you the locations you want to scale.

Simply doing numbers[mask] *= 2 should do the trick :)

Upvotes: 1

Space Impact
Space Impact

Reputation: 13255

Directly assign masked values multiplied by 2 to original array as:

numbers = np.array([-3, 5, 2, -1, -15, 10])
mask = np.abs(numbers) > 5

numbers[mask] = numbers[mask]*2
numbers

array([ -3,   5,   2,  -1, -30,  20])

Upvotes: 1

Related Questions